Hacker News Reader

Comment by melling | Hacker News Reader

Now we just need to make it affordable.

Cost overruns seem to be a big problem in the industry. The US recently canceled 2 plants, and we’re struggling to finish the other 2.

View on HN

Building it makes it affordable. Giant one-off projects have endemic problems with perverse bidding incentives, cost overruns, and engineering problems. Economy of scale production absolutely dominates most industrial supply chains and engineering.

The US and other modern economies got the political will together to subsidize solar when it rarely made sense on a MW/$ basis and now it's dropped to by far the cheapest source (if you can live with intermittent power).

What drives nuclear costs is generally rising construction costs. You can’t bring that down with building more nuclear power plants.

PV got so much cheaper because it’s a mass producible product.

If all the designs were the same you could start prefabbing components, that will do just fine for construction costs.

The reality is that there is an entire supporting industry that needs to be built to make nuclear power plants economical.

Nuclear power plants have special requirements that are rarely met by conventional industry. A lot of nuclear power plant designs involve manufacturing via the largest hydraulic presses we have ever built. There are not many of them and building more of them is expensive so it will raise costs in the short term (10-20 years). The design philosophies have to change considerably if nuclear power wants to be cost effective.

The monolithic nuclear power plant is a dead end.

PV is also cheaper because the possibly polluting production is outsourced to china.

Nuclear is much cheaper when built at scale. Often 3-5x cheaper. Most of the plants built during the nuclear boom in the 1960s and 1970s were built at a cost of ~$2.5 billion per gigawatt of capacity. Note that this capacity is put out 24/7, so it's equivalent to about 4 times as much solar capacity (which typically has a capacity factor of ~25%).

Solar and wind are cheap in terms of watts per day, but become drastically more expensive one you need to start adding storage. Storage costs are typically in the $200-400 per KWh range for battery storage, and $100 - $200 per KWh range for hydroelectric storage, but the latter is geographically limited.

I also haven't really seen anyone discuss the cost of nuclear in human labor costs. No matter what method or design you use you still need educated workers to babysit both the reactor and the standard turbine infrastructure, whose salary dwarfs whatever costs come from fuel or safety countermeasures.

In contrast, solar and wind tend to be very manpower efficient due to the logistical pain of maintaining something spread over 10 sq. miles forcing investments in reliability and autonomy.

Edit: Messed up wording

I'm not quite sure I follow. Doesn't the "logistical pain to maintain something spread over 10 sq. miles" have a human labor cost too?

I meant to imply that the pain of such caused those technologies to invest in autonomy to mitigate needless human labor, but I lost the point when revising the comment. I'll fix it.

Labor costs have to be figured into $/KwH by convention right? They’re not zero for solar.

Ian the same argument for renewables? Subsidize solar panels to make them affordable... the only difference is, solar panels are smaller so we can iterate faster. Maybe that’s what we need to improve about nuclear...

Lots of Small Modular Reactors waiting for political will and funding around. Personally I'm really interested in Soviet-originated SVBR series, but politically it's even more problematic than other designs right now :-(

Let's move it out from government oversight. Anything gov touches gets super expensive, especially in the US. Best example is the Obamacare website.

https://www.bloomberg.com/news/articles/2014-09-24/obamacare...

Imagine how much cheaper you could have built that website with the same functionality. Now apply the same mental process to nuclear power plants.

Yes. The government should only set the safety expectations and procedures, and let the market decide how they do it.

Innovation in nuclear power is broken because of the difficulty to access uranium and use it safely, but if we had innovated, it might have been made safe by now. There has been pilot projects such as a sealed 1m3 cuve, which, by sealing it, solved a lot of problems about safety and human operation.

The government could never build an important highly advanced technical project in a hurry; the atomic bomb project should have been put out to the private sector. /S

The incentives were different.

Anything the private sector touches has safety issues. Best example is the Bhopal Gas Tragedy.

https://en.wikipedia.org/wiki/Bhopal_disaster

Imagine how much safer it would have been under govt regulations. Now apply the same mental process to nuclear power plants exploding.

I agree, but the argument is a non-starter here and equally so with the general public.

Reducing regulation is a lot more palatable to people than eliminating it.

Reply on news.ycombinator.com